February Newsletter: Re-examining alternative energy

Posted by Alex Frey (@alexhfrey )

A semi-recurring theme of IvyVest newsletters is the price of energy, which we have written about several times before.

We’re fascinated by the price of energy for two chief reasons. As a raw material in the cost of just about every good produced (since every good has to be made in factories that take electricity to run and transported to market by trucks, which run on gasoline, a derivative of petroleum), energy impacts the price of everything and the profits of just about every company on Earth.

Secondly, the energy markets have had a wild ten years, with major gyrations in price and a near 180 degree shift in outlook, from one of rising prices and talk of “peak ____” to one of falling prices and talk of excess supply. 

Thirdly, energy presents direct investment opportunities of its own, either through purchasing energy-focused stocks in your “sandbox” (“play-money” account) or commodity ETFs.

The first phase of this period brought increased focus and energy (pun intended) into the idea of powering our homes, cars, and factories through renewable processes that do not rely on fossil fuels or carbon-emitting processes. This is alternative energy.

Because of investments made in alternative energy as well as the inexorable advance of technological (particularly silicon-based technologies like solar panels), the cost of renewable energy systems has been coming down rapidly, and some people are saying that renewables have now reached, or soon will reach, cost parity with conventional generation systems. We wanted to investigate these claims. This may seem to be an esoteric subject, but investors may find it interesting for several reasons. First, it has relevance for the future prices of oil, natural gas, and coal. Second, it will have an effect on the future performance of electric utility companies, and third, it may be relevant to an investor who wants to invest in a residential solar system.

The Methodology of Cost Comparison

Comparing the cost of alternative energies to the price of conventional energy sources may seem to be simple, but it is actually quite complex. Different generation systems have different upfront costs, maintenance costs, fuel costs, and system lifetimes. Aggregating these together into a comparable cost per unit of energy is not simple. 

Further, some are dispatchable, meaning that the output can be adjusted to the demand, and others (particularly renewables) aren’t. Think of solar as an example: it can help alleviate demand in the hot part of a day, but a utility company simply cannot make the sun shine harder when it is hit with a surge in energy demand. But it can turn on a natural gas “peaker” plant.

For these reasons and more, making meaningful cost comparisons is not trivial. Analysts frequently discuss the levelized cost of electricity (LCOE). The LCOE is the cost that a utility would have to charge for electric power over the live time of a facility to just break even. It is computed by considering the lifetime cost of a facility and dividing by the lifetime production of the facility. However, both the costs and the production must be discounted to account for the time value of money, so upfront costs (construction and manufacturing costs) have a bigger effect on the total cost than do maintenance or fuel costs that are incurred in later years. Likewise, production in the early years is more valuable than production in the later years. A discount rate must be used to account for the time value of money, and the rate chosen can have a big effect on the relative cost of different systems.

Likewise, one must estimate future fuel costs, and these can also have a big effect on the LCOE. When computing the LCOE for renewable power sources (in particular, wind and solar), the quality and persistence of the energy source must be considered. For solar, areas closer to the equator with more sunny days are obviously better. The capacity factor (the average output of the system as a fraction of its rated output) is an important parameter in the computation of the LCOE. For wind, the strength and persistence of the wind is important, and the location of the wind farm relative to the market for the electricity is also important. Some of the best wind areas are far from areas of high electrical demand.

There are several competent and unbiased organizations that have computed LCOE values, and we will discuss these shortly. However, there is one more complication to consider. The LCOE, as it is usually computed, is the cost of power at the generating station. It ignores the costs of sending that power to its end source (distribution costs), and it also ignores the cost that a utility must incur to have power available for all foreseeable demands even if it is not generating any power currently (e.g. the cost of building and maintaining that natural gas peaker plant even if it is not online much of the time, because the utility company cannot afford to not provide power when demand spikes). One should really look at the cost of the whole system, including any back up generation or energy storage systems that may be needed to ensure that a utility can meet the demands that are placed on it. 

There is a cost to having a conventional power plant sitting idle but ready, and there is a significant cost to energy storage systems. These costs should really be included in any estimates of the total system cost, but in most cases they are not considered.

How much standby power or storage is needed depends on how much renewable power is included in the generation mix and on how well renewable power availability matches the power demand cycle. In some cases, solar availability may match well with the demand (for instance, in hot desert areas where air conditioning creates the maximum demand in midafternoon). In other cases, particularly with wind, maximum availability may correlate poorly with maximum demand. The need for backup systems may be quite different in these cases. When considering renewable technologies, we think there is probably an optimal mix of renewable generation, dispatchable conventional generation, and energy storage that meets utility needs with minimum system LCOE. However, to our knowledge, this problem has not been addressed, and the optimum mix would probably be very dependent on the exact location of the system.

The Results

In Figure 1, we show LCOE values computed by three agencies that we judge to be competent and unbiased. They are the International Energy Agency (IEA), Lazard Ltd. (a financial advisory and management firm), and the U.S. Energy Information Agency (EIA). Their approaches were not identical, and they used different assumptions in some cases. IEA computed figures for many different countries, but the figures given here were for the United States. The figures are based on actual systems that were recently installed. The EIA averaged data from different U.S. regions. The EIA and the IEA computed figures for plants entering into service in 2022 or 2020. We think that Lazard’s figures are for plants entering service in 2016, but we’re not sure about this. Lazard gives a range of costs for each technology, and we have shown the high and low end of that range. EIA and IEA also gave a range in some cases, and where applicable, we chose a data point in the middle of the range. Given the differences, between the studies, we would not expect that their figures would be in complete agreement, but in fact the three studies agree fairly well on the relative costs of different generation technologies.

The cheapest technology is onshore (land based) wind (off shore wind is not shown in Figure 1, but all three studies say that it is much more expensive). Natural gas combined cycle and utility scalar solar are nearly tied for second place, and the other technologies are more expensive. (Natural gas combined cycle is a technology where the natural gas drives a gas turbine, and the exhaust from the gas turbine is used to power a steam turbine). The solar technology represented here is photovoltaic (silicon chips or thin film). It is interesting that the two renewable technologies are very cost competitive in these analyses. It is also interesting that natural gas is cheaper than coal, even without any CO2 mitigation costs. This doesn’t bode well for efforts to revive the U.S. coal industry.


Figure 1: The levelized cost of electricity as computed by three different studies for different technologies.

The costs given are for the generation of power and do not include distribution costs or CO2 mitigation costs. Also, for renewable technologies (wind and solar), they do not include any other equipment which may be necessary to meet the load demand when the renewables are not available. Data was taken from references 1 to 3.

The cost data in figure 1 is for the generation of electricity only. It doesn’t include distribution or administrative costs, and for wind and solar, it doesn’t include any equipment which may be necessary to meet the load demand when wind or solar is not available. It is difficult to find cost data for a complete system including the necessary standby conventional generation capability or storage capability. Lazard did give one data point for a utility scale solar system with backup battery storage. The system represented a situation in the Southwest United States, but they didn’t give details about how it was designed. For this one case, the LCOE increased from a range of 46 to 56 $/MWh without storage to a value of 92 $/MWh with storage. Thus, the system cost nearly doubled. Reference 4 also computed an LCOE for a system including photovoltaic solar and battery backup. It wasn’t clear how the amount of battery backup was chosen, but including the batteries increased the LCOE for the system by approximately 60%. Based on these data points, it appears that wind and solar may not yet be fully competitive with natural gas systems, but they are close. Of course, using a natural gas plant as a backup for a wind or solar plant might give an attractive system level LCOE with greatly reduced carbon emissions.

The relatively high LCOE value given in figure 1 for residential solar may seem discouraging, but the case for residential solar is actually much better than it appears in figure 1. The costs in figure 1 do not include distribution and administrative costs, which can nearly double the cost of electricity. Power that is generated on the premises does not have to pay these fees. Further, in some localities “net metering” essentially allows residential solar systems to use the grid for free. These localities, when solar customers have surplus power, are allowed to sell it to the utility at the retail price (which includes delivery charges). In essence, net metering allows the residential system to use the electric grid as a perfectly efficient storage battery that stores excess energy and allows the solar customer to take it back when solar power isn’t available. While utility companies find this very objectionable (it costs money to maintain the grid), it is a great benefit to residential solar systems. Also, in many localities, state and local governments give tax subsidies to residential solar systems, and these can significantly reduce the cost of a system. Finally, some utilities are using time of use (TOU) pricing. With TOU pricing, the price for power is higher during times of peak use, such as summer afternoons. This is also when solar is most available, so a solar system may permit the user to avoid buying the high priced grid power at these times. It is beyond the scope of this newsletter to calculate the return that a homeowner might expect from a solar system. An assessment of residential rooftop solar depends very much on exactly where the system is installed and on local utility rules and government subsidies. However, we are persuaded by a number of sources that home solar systems often provide a good return on investment.


The real conclusion from this newsletter is that renewable power is approaching cost parity with conventional generation. It is going to increase in importance regardless of any actions taken or not taken by the EPA. Likewise, coal is likely to continue its decline regardless of any changes in EPA policies. From an environmental perspective, this is a good thing. However, we expect that there will continue to be a growing market for natural gas both as a primary source of electrical power and a backup system for renewables.

1. Lazard LTD, Lazard’s Levelized Cost of Energy Analysis- Version 10.0, 2016.
2. U.S. Energy Information Administration, Levelized Cost and Levelized Avoided Cost of New Generation Resources in the Annual Energy Outlook 2016.
3. International Energy Agency, Projected Costs of Generating Electricity, 2015 Edition.

4. Chun Sing Lai and M.D. McCulloch, Department of Engineering Science, University of Oxford, UK, Levelized Cost of Energy for PV and Grid Scale Energy Storage Systems

Get our next article delivered to your inbox.

Sign up below and be the first to know about our freshest data-driven thinking on the markets, and investing. We will send you no more than one email a week. This is free.

Ready to start putting this into action?

Take a free two-week trial to IvyVest premium -- our premium subscription service. You'll get access to our rules-based dynamic asset allocation model, tools that will show you exactly what you need to buy in your own discount brokerage account (and when to re-balance) to implement it for yourself, and an insightful monthly newsletter that will keep you on abreast of the most important things going on in the markets. There is no credit card required. Get Started Now!

By Alex Frey